paddlets.models.classify.dl.inception_time

class InceptionTimeClassifier(loss_fn: ~typing.Callable[[...], ~paddle.Tensor] = <function mse_loss>, optimizer_fn: ~typing.Callable[[...], ~paddle.optimizer.optimizer.Optimizer] = <class 'paddle.optimizer.adam.Adam'>, optimizer_params: ~typing.Dict[str, ~typing.Any] = {'learning_rate': 0.001}, eval_metrics: ~typing.List[str] = [], callbacks: ~typing.List[~paddlets.models.common.callbacks.callbacks.Callback] = [], batch_size: int = 32, max_epochs: int = 100, verbose: int = 1, patience: int = 10, seed: ~typing.Union[None, int] = None, activation: str = 'ReLU', kernel_size=40, block_out_size=128, block_depth=6, use_bottleneck=True, use_residual=True)[源代码]

基类：PaddleBaseClassifier

InceptionTime是在2019年提出的基于CNN网络的时序分类模型，它的灵感来自于Inception-v4架构

[1] Hassan I.F, et al. “InceptionTime: Finding AlexNet for Time Series Classification”, https://arxiv.org/pdf/1909.04939v3.pdf

参数

  • optimizer_fn (Callable[..., Optimizer]) – 优化算法

  • optimizer_params (Dict[str, Any]) – 优化算法参数

  • eval_metrics (List[str]) – 评估指标

  • callbacks (List[Callback]) – callback方程

  • batch_size (int) – 每个batch的样本量

  • max_epochs (int) – 最大训练轮次

  • verbose (int) – 是否开启日志

  • patience (int) – 训练结束之前等待提升的轮次

  • seed (int|None) – 随机种子

  • activation (str) – 激活方程，默认使用ReLU

  • kernel_size (int) – 卷积核大小，默认为40

  • block_out_size (int) – inception block的输出维度，默认设置为128

  • block_depth (int) – inception block的深度，默认设置为6

  • use_bottleneck (bool) – 是否增加残差

  • use_residual (bool) – 是否启用瓶颈层